In addition to storing images internally, modern ultrasound imaging systems need to be able to transfer images to various types of remote devices via a communications network. To successfully transfer images, the relevant networking features of the ultrasound imager must be compatible with the networking features of the destination remote device. In particular, the ultrasound imager must place the data to be transferred in a format which can be handled by the destination remote device. An attempt to accomplish the foregoing is the adoption of the DICOM (Digital Imaging and Communications in Medicine) standards, which specify the conformance requirements for the relevant networking features. The DICOM standards are intended for use in communicating medical digital images among printers, workstations, acquisition modules (such as an ultrasound imaging system) and file servers. The acquisition module is programmed to transfer data in a format which complies with the DICOM standards, while the receiving device is programmed to receive data which has been formatted in compliance with those same DICOM standards.
DICOM involves more than digital image transfer. DICOM functionality includes the following Service Classes: archive/transfer images: store (across network); archive/interchange images: media storage; query for information and retrieve images; make image hard copies: print management; patient, study and results management; radiology information system modality: worklist management; and test connectivity: verification. A fundamental concept employed in DICOM is “Services on Objects”. One example of an “Object” is an ultrasound image. Two examples of a “Service” are the “Store” and “Query/Retrieve” functions. In DICOM, methods of operating on information objects are referred to as “Service Object Pair Classes” (SOP Classes). Examples of SOP Classes are “Store an ultrasound image”, “Print an ultrasound image”, “Find which studies there are for a certain patient”, “Retrieve all studies of a certain patient” and “Retrieve a worklist”. Unique Identifiers (UIDs) are defined for all SOP Classes. UIDs are also given to all studies, series and images. These UIDs are, for instance, used for retrieval. In the DICOM vernacular, a patient has a study which comprises a study component, e.g., examination using a particular modality. Images acquired in sequence in the course of a study on a patient form a series of objects.
The DICOM system is based on the client/server concept. The device which uses a service (on objects) is the client device, while the device which provides the service is the server device. The client device is referred to as a Service Class User (SCU), while the server device is referred to as a Service Class Provider (SCP). The SCU sends a Service Request to the SCP over a local area network (LAN). The SCP sends back a response to the SCU over the same LAN. If the response is affirmative and a communications syntax is agreed upon, an association between the SCU and the SCP is opened and data can be transferred between the two devices. In the DICOM system a device is not limited to one role: it can be both SCU and SCP at different times.
The DICOM system is designed to facilitate the communication of digital images of different types, e.g., X-ray, computerized tomography, magnetic resonance and ultrasound imaging. In an ultrasound imager having conventional DICOM capability, three local real-world activities occur: Image Send, Image Print and Remote Verification. Image Send and Image Print can be done in either automatic or manual mode. Verification of remote DICOM devices configured on the ultrasound imager is performed when the imager is powered up or when requested by the system operator.
All DICOM activities are handled in a queued manner by application software running on a host computer incorporated in the imager. In one type of ultrasound imager, the user can select any image in cine memory to be sent in DICOM format via a LAN to a remote device having DICOM capability. The host computer of the ultrasound imaging system is programmed with DICOM system software which facilitates transmission of image frames from the cine memory to the remote DICOM device via the host computer hard disk and the LAN.
In the conventional ultrasound imager, Image Send can be used in automatic or manual mode, depending on the user configuration. When automatic mode is configured, console keys are used to capture the image and to store it on the hard disk. The request is queued to a DICOM queue manager (preferably implemented in software), which requests an association with the destination remote device. After the association with the remote device has been opened, the queue manager “pushes” the image to the remote device without user intervention. The transfer is done in the background while scanning or other operator activities continue. In manual mode, the captured images are archived on the hard disk or on a MOD during the exam(s). Upon completion of the exam(s) the images are tagged using an archive menu and queued to any of the network devices that have been configured on the imager. The images are sent sequentially in the background while scanning or other operator activities proceed. Image Print works much the same way as Image Send, in both automatic and manual modes, the only difference being that the destination device is a printer.
In order to accomplish image transfer, the ultrasound imaging system must know the configuration of the destination remote device prior to attempting to communicate with that device. The configuration data for the destination remote device is typically inputted to the ultrasound imager during software installation by a field engineer, although the DICOM network can be configured at any time. When the imager receives an instruction to transmit data to a particular remote device from the system operator, the imager software converts the image data to be transferred into the DICOM format required by the destination remote device, based on the configuration data for that device stored in system memory. The imager also sends a request over the network to the destination remote device to open an association, i.e., to connect the imager to the destination remote device. If the remote device responds in the affirmative, the imager and remote device then agree on which SOP Class is to be used and which device will act as the server and which as the client. The ultrasound imager also selects the appropriate encoding syntax from those accepted by the remote device. Other communication parameters are also negotiated.
After the DICOM communications protocol has been settled, the association is opened and the imager attempts to send the DICOM-formatted image frame (object) to the remote device via the network. The transfer is done in the background while scanning or other operator activities continue. If the remote device is a storage device, each image frame is transferred singly in response to a Send request inputted by the operator., The conventional imager with DICOM capability will open an association with a storage device in response to each “send to a storage device” instruction. If a transfer is successful, the association for that transfer is immediately closed. If the remote device is a printer configured to print multi-image film, then a number of images are accumulated to make up a multi-image film and an association is opened in response to a Send instruction when a number of images sufficient to fill the multi-image film have been accumulated. After the full film session of images has been transmitted, the association between the imager and printer is closed.
If the destination remote device sends back a message indicating successful receipt of the transmitted data, the DICOM-formatted image frame can be deleted from the imager memory. Alternatively, the system operator can instruct the imager to retain the DICOM-formatted image frame on the imager hard disk or to store it on a MOD inserted in the imager.
In addition to the digitized image (i.e., pixel data), the DICOM object transferred from the ultrasound imager also includes attribute information. For example, the attribute information may include patient attributes (e.g., patient name and patient identification number), study attributes (e.g., accession number and study date), series attributes (e.g., modality type and series date), and image attributes (e.g., image type and numbers of rows and columns). Each attribute has a name, a value representation and a tag. A tag is a four-byte number (e.g., in hexadecimal code) unique to the attribute. The value representation defines what type of value it can have (e.g., a 64-character string, binary data, etc.).
In accordance with DICOM standards, there are three types of attributes. Type 1 comprises attributes which are mandatory and must always be present with a value; Type 2 comprises attributes which are mandatory but are allowed to be empty; and Type 3 comprises attributes which are optional and are also allowed to be empty. An incompatibility between two devices may arise, for example, if the receiving device requires that a Type 3 attribute be transmitted while the sending device does not include that attribute in its transmission. As a result, even if both devices are configured in accordance with current DICOM standards, the data transfer cannot occur. Thus, even mutual conformance to DICOM standards does not guarantee that two devices can be compatibly connected to each other.
In accordance with a further aspect of the DICOM system as currently implemented, an ultrasound imaging system can retrieve a worklist from a Radiology Information System (RIS) at a hospital via the LAN. The retrieved worklist may, e.g., comprise all patients to be examined on a particular day using that particular ultrasound imager. The worklist includes the following information for each patient: name, identification number, sex, birth date, accession number, study data, etc. The information retrieval is initiated by the ultrasound imager. In response to this query, the RIS transmits the worklist to the ultrasound imager, which stores it in memory. This worklist is then available for viewing by the sonographer. The patient currently being examined can be selected from the worklist.
Because the DICOM capability is implemented in software, these features of the ultrasound imaging system can be readily upgraded. One goal of such upgrades is to increase the efficiency of the system operator by making the system simpler to operate, e.g., by requiring fewer manipulations to activate a particular operation. Another goal of system upgrades is to increase the ability of the imager to connect rapidly, efficiently and reliably to remote devices on the network, i.e., to increase connectivity.